Week 10 DVOA Ratings

by Aaron Schatz

The biggest development of Week 10, of course, is the tsunami of injuries that has hit the Indianapolis Colts. The Colts remain number two for 2007, but they are just barely ahead of the Cowboys, and the Patriots and Colts are no longer the two best teams of the DVOA era. The Colts have fallen back into the pack, although they are still the sixth-best team in history. (They've fallen behind Dallas and into third place in WEIGHTED DVOA.)

The other big news of the week is Green Bay's destruction of Minnesota, which moves the Packers up into fifth place. I think logically we would all agree this makes sense. The five teams that are 7-2 or better are now the top five teams in DVOA, and all five are in the Top 10 for both offense and defense. The Vikings, meanwhile, have been jerked around wildly because of two games at opposite sides of the spectrum. The Week 9 win over San Diego currently has a DVOA of 101.2%, which is the fourth-best single game of 2007. The Week 10 loss has a DVOA of -71.4% -- that doesn't qualify for the ten worst losses of the year, but it is pretty close. So the Vikings have gone from 16 to 8 to 15 in the space of two weeks. (Last week's rating actually was artificially high due to an error in special teams, but even the corrected Vikings were eighth overall.)

Oh, and because I know you will ask, the other top single games of the year, based on current opponent adjustments, are Arizona's upset of Pittsburgh, New England's demolition of Washington, and two Pittsburgh games: Week 1 against Cleveland and Week 9 against Baltimore. Let's hit the numbers...

* * * * *

Here are the Football Outsiders team efficiency ratings through 10 weeks of 2007, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here [1].) OFFENSE and DEFENSE DVOA are adjusted for opponent and consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.

As of this week, opponent adjustments are at full strength, although they will continue to change throughout the season as we get more information on how good each team really is.

To save people some time, we request that you please use the following format for all complaints:

<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>

	TEAM	TOTAL DVOA	LAST WEEK	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
1	NE	73.2%	1	68.6%	9-0	48.5%	1	-18.5%	3	6.3%	5
2	IND	42.1%	2	34.6%	7-2	30.6%	2	-17.4%	4	-6.0%	31
3	DAL	41.6%	3	39.0%	8-1	28.3%	3	-12.8%	6	0.5%	14
4	PIT	32.8%	4	39.1%	7-2	12.6%	9	-23.8%	2	-3.6%	25
5	GB	23.7%	7	28.3%	8-1	17.0%	4	-6.4%	9	0.2%	15
6	TB	18.1%	6	13.1%	5-4	15.5%	6	-4.5%	12	-2.0%	18
7	SD	14.7%	13	8.5%	5-4	0.3%	16	-5.1%	11	9.3%	1
8	JAC	13.8%	9	11.0%	6-3	10.6%	10	0.5%	15	3.7%	9
9	SEA	12.1%	10	21.9%	5-4	-1.1%	18	-5.5%	10	7.7%	2
10	TEN	11.2%	5	11.6%	6-3	-11.0%	25	-25.6%	1	-3.5%	24
11	NYG	6.7%	11	9.8%	6-3	7.6%	12	-3.8%	13	-4.7%	29
12	CLE	5.3%	15	-0.5%	5-4	13.3%	8	15.3%	27	7.3%	3
13	PHI	4.9%	12	2.9%	4-5	15.9%	5	8.1%	22	-2.9%	22
14	WAS	3.3%	14	-4.7%	5-4	-4.9%	22	-7.4%	8	0.9%	13
15	MIN	2.6%	8	0.9%	3-6	2.5%	15	1.6%	16	1.7%	11
16	CIN	-0.3%	17	3.5%	3-6	15.5%	7	14.9%	26	-0.9%	16
	TEAM	TOTAL DVOA	LAST WEEK	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
17	BUF	-2.4%	16	-6.1%	5-4	-9.1%	23	0.0%	14	6.7%	4
18	ARI	-3.3%	21	4.5%	4-5	-2.8%	19	1.8%	17	1.4%	12
19	DEN	-6.0%	24	-8.3%	4-5	6.3%	13	11.1%	23	-1.2%	17
20	BAL	-9.8%	20	-5.4%	4-5	-22.6%	29	-15.4%	5	-2.6%	21
21	NO	-12.4%	19	-13.0%	4-5	8.7%	11	16.6%	29	-4.6%	28
22	DET	-13.8%	18	-4.1%	6-3	-4.0%	20	3.5%	20	-6.2%	32
23	HOU	-14.4%	23	-11.7%	4-5	-0.8%	17	15.3%	28	1.8%	10
24	KC	-16.4%	22	-11.8%	4-5	-22.6%	30	-8.2%	7	-2.0%	19
25	ATL	-17.6%	26	-7.4%	3-6	-10.1%	24	5.1%	21	-2.4%	20
26	MIA	-19.8%	28	-28.5%	0-9	2.5%	14	18.6%	31	-3.7%	26
27	CAR	-20.4%	25	-15.1%	4-5	-12.9%	26	2.1%	18	-5.4%	30
28	CHI	-20.7%	27	-17.2%	4-5	-24.0%	31	2.1%	19	5.4%	7
29	NYJ	-25.6%	29	-27.9%	1-8	-4.8%	21	26.9%	32	6.2%	6
30	STL	-34.0%	31	-34.9%	1-8	-16.7%	27	13.4%	25	-4.0%	27
31	OAK	-35.3%	30	-28.1%	2-7	-19.1%	28	13.1%	24	-3.1%	23
32	SF	-51.6%	32	-46.0%	2-7	-38.0%	32	17.8%	30	4.2%	8

ESTIMATED WINS [2] uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close. It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles.
PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
VARIANCE [3] measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from least consistent (#1, highest variance) to most consistent (#32, smallest variance).

	TEAM	TOTAL DVOA	W-L	ESTIM. WINS	RANK	WEIGHTED DVOA	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
1	NE	73.2%	9-0	9.0	1	73.9%	1	5.8%	5	-2.3%	19	2.7%	32
2	IND	42.1%	7-2	8.5	2	40.3%	3	7.8%	1	-11.4%	26	10.4%	15
3	DAL	41.6%	8-1	7.2	3	41.8%	2	1.5%	14	-4.1%	20	8.1%	22
4	PIT	32.8%	7-2	6.0	5	29.0%	4	-5.1%	26	-0.5%	15	31.6%	2
5	GB	23.7%	8-1	6.4	4	23.7%	5	-1.0%	18	-16.4%	27	8.4%	21
6	TB	18.1%	5-4	5.7	7	18.2%	6	-0.5%	17	-21.8%	32	21.9%	6
7	SD	14.7%	5-4	5.2	9	17.0%	7	4.7%	10	-9.4%	23	30.6%	3
8	JAC	13.8%	6-3	5.9	6	14.3%	8	1.6%	13	2.9%	14	7.1%	23
9	SEA	12.1%	5-4	5.1	11	11.2%	9	-9.7%	32	-17.1%	29	13.4%	11
10	TEN	11.2%	6-3	5.2	10	11.1%	10	-1.2%	19	-1.0%	17	8.4%	20
11	NYG	6.7%	6-3	5.0	14	8.5%	11	0.0%	16	7.5%	10	5.7%	28
12	CLE	5.3%	5-4	5.2	8	7.6%	12	5.2%	8	-18.0%	30	9.0%	19
13	PHI	4.9%	4-5	5.1	12	3.0%	13	1.9%	12	16.5%	3	17.6%	8
14	WAS	3.3%	5-4	4.6	17	2.9%	14	5.1%	9	14.3%	6	9.7%	17
15	MIN	2.6%	3-6	4.9	15	2.2%	15	3.8%	11	-19.9%	31	23.3%	5
16	CIN	-0.3%	3-6	5.1	13	-0.6%	17	5.9%	4	-9.9%	24	3.8%	31
	TEAM	TOTAL DVOA	W-L	ESTIM. WINS	RANK	WEIGHTED DVOA	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
17	BUF	-2.4%	5-4	4.7	16	1.0%	16	6.0%	2	14.6%	5	10.8%	14
18	ARI	-3.3%	4-5	3.8	21	-3.4%	18	-6.3%	29	-16.4%	28	33.3%	1
19	DEN	-6.0%	4-5	4.3	18	-8.1%	19	5.9%	3	-10.0%	25	19.2%	7
20	BAL	-9.8%	4-5	3.5	23	-10.5%	21	-8.0%	30	26.7%	1	6.4%	26
21	NO	-12.4%	4-5	4.0	20	-8.7%	20	-2.7%	23	-9.2%	22	10.1%	16
22	DET	-13.8%	6-3	4.2	19	-14.2%	22	-6.1%	27	16.1%	4	24.4%	4
23	HOU	-14.4%	4-5	3.6	22	-18.0%	25	-2.8%	24	12.0%	9	11.1%	13
24	KC	-16.4%	4-5	3.4	25	-15.0%	23	-2.4%	22	-2.2%	18	9.5%	18
25	ATL	-17.6%	3-6	3.3	26	-16.4%	24	-8.5%	31	6.7%	12	6.7%	25
26	MIA	-19.8%	0-9	2.9	28	-21.1%	28	5.2%	7	12.1%	8	5.9%	27
27	CAR	-20.4%	4-5	3.4	24	-19.2%	26	-2.8%	25	7.5%	11	14.0%	10
28	CHI	-22.5%	4-5	3.3	27	-20.2%	27	0.8%	15	5.0%	13	4.7%	30
29	NYJ	-25.9%	1-8	2.6	29	-24.0%	29	5.3%	6	21.3%	2	5.1%	29
30	STL	-34.0%	1-8	1.9	30	-33.5%	30	-2.1%	21	-0.7%	16	14.3%	9
31	OAK	-35.3%	2-7	1.6	31	-34.1%	31	-6.2%	28	12.4%	7	12.1%	12
32	SF	-51.6%	2-7	1.1	32	-53.4%	32	-1.4%	20	-5.4%	21	7.0%	24

Our feature on the best teams of the DVOA era changes again this week. First of all, we'll say goodbye to our "best defensive teams" table. Tennessee is still number one for 2007 after their loss to Jacksonville, but they are no longer near the top teams since 1996. However, and it really pains me to have to bring this up, San Francisco has finally reached the level of incompetence that forces us to look both ways: the best and worst DVOA ratings ever.

Best and Worst DVOA Ever Watch

BEST TOTAL DVOA AFTER WEEK 10			BEST OFFENSIVE DVOA AFTER WEEK 10			WORST TOTAL DVOA AFTER WEEK 10			WORST OFFENSIVE DVOA AFTER WEEK 10
2007	NE	73.2%	2007	NE	48.5%	2005	SF	-63.0%	2005	SF	-48.2%
1996	GB	48.6%	2004	IND	41.2%	2000	CIN	-54.4%	1996	STL	-39.7%
2001	STL	47.5%	2000	STL	39.7%	2007	SF	-51.6%	2002	HOU	-39.3%
2001	PHI	46.2%	2000	IND	38.1%	1996	STL	-49.8%	1999	ARI	-38.3%
1999	STL	45.7%	2006	IND	34.4%	2005	HOU	-48.4%	2007	SF	-38.0%
2007	IND	42.1%	1999	WAS	31.2%	2002	HOU	-47.2%	2006	OAK	-36.9%
2000	TEN	41.8%	2001	STL	31.0%	1999	CLE	-47.2%	1997	NO	-35.2%
2007	DAL	41.6%	1998	DEN	30.7%	2000	CLE	-47.2%	2004	MIA	-35.0%
2002	TB	41.6%	2007	IND	30.6%	2000	ARI	-45.5%	1999	SD	-34.0%
1998	DEN	38.7%	2002	KC	30.3%	1999	CIN	-42.7%	2000	CLE	-33.8%

Yes, they are that bad, although they are not as bad as... themselves, two years ago. I doubt that this year's 49ers are going to get down to that level of ineptitude. The 49ers have pulled away from the bottom of the pack, but it really doesn't compare to the way the Patriots have pulled away from the top of the pack. Two teams from the "DVOA Era" are listed in Eddie Epstein's book on the best NFL teams ever. He ends up ranking the 1999 Rams tenth and the 1996 Packers sixth. Notice how much the Patriots are blowing away those two dominant teams. Also notice that they are the most consistent team in the league this year. Wow.

Speaking of players who make you say "Wow," how about that Devin Hester. It's been a very popular question around here: How is Chicago's DVOA affected by teams trying short kicks or squibs to avoid Hester. The short answer, prior to this week, was "about two percentage points."

The long answer actually addresses three questions:

1) Are teams treating Hester differently compared to the other good kick returners of recent years?
2) Is this a winning or losing strategy for Chicago's opponents?
3) If the answer to number one is "yes," how does that affect DVOA?

To answer the first question, take a look at this table showing the teams with the lowest average gross kickoff distance AGAINST, since the introduction of the K-ball in 1999.

Year	Team	Avg Kick	Main Returner
2007	CHI	56.5	Devin Hester
2000	CIN	58.5	Tremain Mack
2001	CLE	59.8	Ben Gay, Andre King
2004	CLE	59.9	Richard Alston
2003	NE	60.2	Bethel Johnson
2001	NYJ	60.3	Craig Yeast, Chad Morton
2005	CLE	60.4	Josh Cribbs
2003	PIT	60.4	Ike Taylor, Antwaan Randle El
2001	CHI	60.5	Autry Denson, Leon Johnson
2004	KC	60.5	Dante Hall

OK, so Hester is truly unique. The average kickoff to the Bears this year is shorter than to any other team in the last eight years, and there are extenuating circumstances with the other teams below 60.0.

The 2000 Bengals are just weird, because Tremain Mack was one of the worst kick returners in the league that year. It looks like the Bengals were getting blown out so often that teams were constantly kicking short in the second half, to prevent the rare touchdown return that would launch the surprise comeback.

There is clearly something going on with Cleveland that I never noticed before. Two other Cleveland years just miss this list: 1999 (60.5 -- it misses by hundredths of a yard) and 2006 (61.0). A stadium near the water affects San Francisco's punting numbers, and I have an adjustment for that. Perhaps the winds affect kickoffs in Cleveland rather than punts, and I need to add an adjustment for that too.

It turns out teams didn't really treat Dante Hall the way they are treating Hester. Even in 2003, the year Hall was allegedly an MVP candidate, the average kickoff to Kansas City went 63.4 yards. In fact, the average kickoff distance in Weeks 1-8 (63.4) was no different from the average kickoff distance in Weeks 10-17 (63.3), after Hall had established himself as a major threat.

OK, second question: Is this a good idea? I just looked at this for an upcoming issue of ESPN the Magazine, and here's what I found.

Hester did not take over as Chicago's main kickoff returner until Week 13 of last year, and teams generally kicked to him during last year's regular season and for the first four weeks of this year. Overall, from Week 13 of 2006 through Week 8 of 2007 teams kicked away to Hester 46 times and tried to avoid him 35 times. (This is my best estimate, obviously I may be wrong about the intention of a couple of those kicks.)

The net gain from kicking to Hester averaged 37.8 yards. The net gain from avoiding Hester averaged just 34.0 yards.
Chicago's averaging starting position was their own 31.7-yard line when the other team kicked to Hester, and their own 36.9-yard line when the other team avoided Hester.
On average, kicking deep to Hester was worth -.29 points of field position compared to the league average for kickoffs. The average kickoff which avoided Hester was worth -.42 points of field position.

The stats for kicking deep to Hester would be even better if we included this week's game against Oakland. The numbers seem to say that it is better to kick to Hester, and don't even consider two other issues: fumbles and penalties. Hester may be Mr. 100 Speed, but the guy has butterfingers. Four fumbles is an astonishing number for a guy who has only fielded 50 kickoffs in his career. Two of those fumbles were recovered by the kicking team, giving them the ball deep in Chicago territory. (On punts, the problem is even worse: 11 muffs or fumbles in a year and a half, twice as many as any other player in the NFL.)

Every great Hester run is also one block in the back away from disappearing into thin air. Hester lost a touchdown to a penalty against Seattle in the playoffs, and he lost another one against Kansas City in Week 2. Even a shorter Hester return could be docked 10 yards on a penalty, and that's not going to happen when Israel Idonije is fair catching a short kick.

Hester is a good kickoff returner, not a great one. He's a great punt returner (despite the muffs) but kickoff returns and punt returns use different skills. Late in the half or nursing a lead, it makes sense to avoid Hester. Otherwise, teams are just handing Chicago the ball 20 yards from field-goal position, and even an inept offense like Chicago will gain 20 yards on a lot of drives. Avoiding Hester is a losing strategy.

However, it is a strategy people are using, which means that we need to start accounting for it. The Bears seem to be an exception to the general rule that the receiving team has no control over opposing kickoffs. Therefore, beginning this week, the 2007 special teams ratings include an "avoid Hester" penalty. In general, squib kicks at the end of a half are left out of DVOA. If Stephen Gostkowski squibs it 30 yards with five seconds left in the first half, that doesn't really say anything about Gostkowski's kicking ability. The squibs to Hester which came earlier in the game were being counted as bad kicks for the kicking team but were showing up in Chicago's ratings as very short returns.

Now, based on my subjective decision, I've also coded a number of kickoffs to Chicago as "Hester squibs." On these kicks, I will be giving Chicago value based not on the return of the kickoff, but the net value of the kickoff compared to all NFL kickoffs. It's a temporary solution, and it is not exact, and I'm sure people can criticize it and pick it apart, but it seems to work for now. Overall, this raises Chicago's special teams DVOA from 3.5% to 5.4%, and the Bears go from 23rd in kick returns to 11th. I also marked a couple of Leon Washington returns as "Hester squibs," and looked for them for other guys (Cribbs, for example) but couldn't find anything that clearly qualified.

In the off-season, we can go through and look to see if there is a pattern to squib kicks in general, and figure out how to handle short kicks of all kinds: the end-of-half squibs, the "we have a 28-point lead so who cares" squibs, and the "Devin Hester scares me" squibs.

Housekeeping: The premium database [4] is updated, the team stats pages are updated, the playoff odds [5] are updated, and the Loser League [6] is updated with the first week of the second contest. Note that the premium data and playoff odds were both computed prior to installing the "Hester fix" to special teams, so numbers may differ slightly from those here. I'll get to updating all the individual stuff tonight when I get some free time.

For (short) comments on every team, remember to look for DVOA on AOL, every Wednesday. (This will be linked on the FO Goes Mainstream [7] page.)